GPT-5.2: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
OpenAIopenai/flex | Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 100%◆ (best) | 1.04s◆ (best) | 1.86s◆ (best) | 115 tps◆ (best) | 400K | $0.875 / $7.00◆ (best) |
OpenAIopenai | Healthy | 99.95% | 99.99% | 99.93% | 99.44% | 1.96s | 13.0s | 51 tps | 400K | $1.75 / $14.00 |
Azureazure | No data | — | 99.96% | 99.97% | 99.89% | 3.32s | 44.3s | 84 tps | 400K | $1.75 / $14.00 |
OpenAIopenai/fast | No data | — | 100% | 100%◆ (best) | 100%◆ (best) | 2.86s | 3.49s | 83 tps | 400K | $3.50 / $28.00 |
OpenAIopenai/priority | No data | — | — | — | — | — | — | — | 400K | $3.50 / $28.00 |
- OpenAI100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 1.04s◆ (best)
- p99
- 1.86s◆ (best)
- Throughput
- 115 tps◆ (best)
- Context
- 400K
- $/Mtok
- $0.875 / $7.00◆ (best)
openai/flex - OpenAI99.95%Healthy24h 99.99%
- 7d
- 99.93%
- 30d
- 99.44%
- p50
- 1.96s
- p99
- 13.0s
- Throughput
- 51 tps
- Context
- 400K
- $/Mtok
- $1.75 / $14.00
openai - No data24h 99.96%
- 7d
- 99.97%
- 30d
- 99.89%
- p50
- 3.32s
- p99
- 44.3s
- Throughput
- 84 tps
- Context
- 400K
- $/Mtok
- $1.75 / $14.00
azure - No data24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 2.86s
- p99
- 3.49s
- Throughput
- 83 tps
- Context
- 400K
- $/Mtok
- $3.50 / $28.00
openai/fast - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 400K
- $/Mtok
- $3.50 / $28.00
openai/priority
Measured 4m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
Availability
Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.
Time to first token
Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window.
Throughput
Median tokens per second.